3 months ago
Toronto, CanadaSenior
Responsibilities
- Embed with internal teams and customers to scope needs and build agents from scratch for production workflows.
- Own deployments end-to-end, including discovery, development, integration, activation, tuning, pilots, demos, and adoption.
- Architect agentic systems with reasoning, planning, tool use, memory, multi-agent coordination, and safety guardrails.
- Build secure tool-use infrastructure across APIs, databases, and services with permissioning, sandboxing, and human-in-the-loop controls.
- Develop SDKs, reusable patterns, and blueprints for rapid agent development and deployment.
- Design evaluations, observability, tracing, and guardrails to measure quality, detect regressions, and manage model and data drift.
- Operate multi-model inference for text, voice, code, and vision while managing latency, throughput, and cost.
- Set technical direction and standards for agentic systems and mentor engineers while partnering with ML, product, and security teams.
Requirements
- At least 5 years of software engineering experience, including at least 3 years in AI systems or LLM applications.
- Production experience shipping systems end-to-end and strong proficiency in Python across pipelines, agents, services, and instrumentation.
- Strong understanding of LLM agent architectures including ReAct, RAG, tool use, and multi-agent systems, with hands-on orchestration and evaluation experience.
- Production experience with AWS, Azure, Docker, and Kubernetes.
- Understanding of agentic security risks including prompt injection, privilege escalation, and data leakage.
- Strong customer and stakeholder judgment, communication skills, and comfort structuring ambiguous problems.
- Preferred experience includes agentic systems in regulated industries, AWS SageMaker, Amazon Bedrock, Azure ML, Azure OpenAI, MCP, agent evaluation and observability tooling, vLLM, TensorRT-LLM, Triton, fine-tuning, quantization, LoRA, Temporal, Airflow, Prefect, voice or multimodal inference, edge inference, and testing non-deterministic AI systems.
Categories
About Paytm
Paytm builds a digital payments and financial services platform for consumers and merchants in India, spanning the Paytm app for UPI, wallet and bill pay, plus QR/POS and Soundbox devices for in-store acceptance. It monetizes through payment processing fees, merchant subscriptions, and commissions on financial products such as lending and insurance. Founded in 2010 and headquartered in Noida, it is publicly listed in India and serves over 300M users and 20M merchants.